CN104020724A - Alarm monitoring method and device - Google Patents
Alarm monitoring method and device Download PDFInfo
- Publication number
- CN104020724A CN104020724A CN201310066956.7A CN201310066956A CN104020724A CN 104020724 A CN104020724 A CN 104020724A CN 201310066956 A CN201310066956 A CN 201310066956A CN 104020724 A CN104020724 A CN 104020724A
- Authority
- CN
- China
- Prior art keywords
- mrow
- sample data
- type
- control
- distribution type
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 115
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000009826 distribution Methods 0.000 claims abstract description 179
- 238000004519 manufacturing process Methods 0.000 claims abstract description 42
- 230000000737 periodic effect Effects 0.000 claims description 25
- 230000002146 bilateral effect Effects 0.000 claims description 11
- 230000008859 change Effects 0.000 claims description 10
- 238000012423 maintenance Methods 0.000 claims description 6
- 238000012806 monitoring device Methods 0.000 claims description 5
- 238000003070 Statistical process control Methods 0.000 abstract description 12
- 230000002159 abnormal effect Effects 0.000 abstract description 6
- 230000008569 process Effects 0.000 description 17
- 238000004458 analytical method Methods 0.000 description 12
- 238000012360 testing method Methods 0.000 description 8
- 239000002904 solvent Substances 0.000 description 5
- 239000000126 substance Substances 0.000 description 5
- 230000001186 cumulative effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000004886 process control Methods 0.000 description 3
- 238000005530 etching Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000003486 chemical etching Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0224—Process history based detection method, e.g. whereby history implies the availability of large amounts of data
- G05B23/0227—Qualitative history assessment, whereby the type of data acted upon, e.g. waveforms, images or patterns, is not relevant, e.g. rule based assessment; if-then decisions
- G05B23/0235—Qualitative history assessment, whereby the type of data acted upon, e.g. waveforms, images or patterns, is not relevant, e.g. rule based assessment; if-then decisions based on a comparison with predetermined threshold or range, e.g. "classical methods", carried out during normal operation; threshold adaptation or choice; when or how to compare with the threshold
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F25—REFRIGERATION OR COOLING; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS; MANUFACTURE OR STORAGE OF ICE; LIQUEFACTION SOLIDIFICATION OF GASES
- F25J—LIQUEFACTION, SOLIDIFICATION OR SEPARATION OF GASES OR GASEOUS OR LIQUEFIED GASEOUS MIXTURES BY PRESSURE AND COLD TREATMENT OR BY BRINGING THEM INTO THE SUPERCRITICAL STATE
- F25J3/00—Processes or apparatus for separating the constituents of gaseous or liquefied gaseous mixtures involving the use of liquefaction or solidification
- F25J3/02—Processes or apparatus for separating the constituents of gaseous or liquefied gaseous mixtures involving the use of liquefaction or solidification by rectification, i.e. by continuous interchange of heat and material between a vapour stream and a liquid stream
- F25J3/04—Processes or apparatus for separating the constituents of gaseous or liquefied gaseous mixtures involving the use of liquefaction or solidification by rectification, i.e. by continuous interchange of heat and material between a vapour stream and a liquid stream for air
- F25J3/044—Processes or apparatus for separating the constituents of gaseous or liquefied gaseous mixtures involving the use of liquefaction or solidification by rectification, i.e. by continuous interchange of heat and material between a vapour stream and a liquid stream for air using a single pressure main column system only
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Automation & Control Theory (AREA)
- Mathematical Physics (AREA)
- Physiology (AREA)
- Cardiology (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Computational Mathematics (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Veterinary Medicine (AREA)
- Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Molecular Biology (AREA)
- Heart & Thoracic Surgery (AREA)
- Evolutionary Biology (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Biophysics (AREA)
- Algebra (AREA)
- Pulmonology (AREA)
- Databases & Information Systems (AREA)
- Thermal Sciences (AREA)
- Mechanical Engineering (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
Abstract
The application discloses an alarm monitoring method and device, wherein the method includes collecting sample data used for alarm monitoring in a production process, determining the distribution pattern of the sample data, obtaining an alarm monitoring strategy corresponding to the distribution pattern, and using the alarm monitoring strategy to perform alarm monitoring on the production process. According to the alarm monitoring method, by selecting control limits corresponding to different distribution patterns, control limits used for alarm monitoring for sample data of different distribution patterns can be rapidly determined, thereby solving the technical problem in the prior art that for large sample data of different distribution patterns, acquisition of accurate control limits needs to take a relatively long time, and thus a statistical process control method can be established quickly so as to lower the false alarm probability and effectively monitor abnormal fluctuations of mass data.
Description
Technical Field
The present application relates to the field of process control, and in particular, to a method and an apparatus for alarm monitoring.
Background
In the production process of the product, the size and other specifications of the product can fluctuate for some reasons, and the fluctuation has a great influence on the quality of the product, but the influence caused by the fluctuation can be completely avoided and eliminated by taking measures, namely process control. Statistical Process Control (SPC) is a Process Control tool by means of mathematical Statistical method, which analyzes and evaluates the production Process, finds the sign of systematic factors in time according to feedback information, and takes measures to eliminate the influence, so that the Process is maintained in a controlled state only influenced by random factors, thereby achieving the purpose of controlling quality.
The implementation of statistical process control is divided into two phases: an analysis phase and a monitoring phase. In the analysis stage, after the production preparation is completed, the control boundary of the control chart is calculated by using a plurality of groups of sample data collected in the past stable production process, the control chart and the histogram for analysis are made, or process capability analysis is carried out, and whether the production process is in a statistical steady state or not and whether the process capability is enough or not are checked. If any one of the analysis control charts can not be met, the reason is found, improvement is carried out, production and analysis are prepared again, the two purposes of the analysis stage are achieved, the analysis stage can declare to be finished, the statistical process control monitoring stage is entered, and the analysis control chart is converted into a control chart. The main work of the monitoring stage is to use a control chart for control to carry out monitoring, wherein the control boundary of the control chart is determined according to the result of the analysis stage, the data of a new production process is also drawn on the control chart in time, the control chart is closely observed, the fluctuation condition of points in the control chart can show that the process is controlled or out of control, and if the out of control is found, the reason is found and the influence is eliminated as soon as possible. Monitoring may be sufficient to enable statistical process control preventive control. In the actual application of the plant, the above two stages must be passed for each control item, and such a process from analysis to monitoring is repeated as necessary.
At present, a statistical process control method is usually based on that process quality characteristic data in a steady state obeys normal distribution, and a standard deviation of n times plus or minus an average value (generally, 3 times of standard deviation is used in the industry) is used as a control boundary for monitoring, so that abnormal fluctuation can be effectively monitored, and the probability of a first type of error (false alarm) is ensured to be about 0.27%.
However, in the production Process of modern industry, there are a lot of Process parameters, for example, in the semiconductor manufacturing industry, there are data regarding product characteristics, such as film thickness, CD, electrical parameters, yield, and also a lot of parameters regarding equipment and production environment, such as dust falling amount, output current (voltage), Process (Process) time, temperature and humidity. It has been found through research that a large amount of data does not follow a normal distribution, and there is also much data that may even be a distribution that cannot be described by a mathematical formula. Thus, the probability of false alarms is greatly increased if a statistical process control method based on normal distribution is applied.
In addition, although there are many non-normal distribution analysis methods in the industry, most of them require professional tools to specifically analyze data, and even after the data is deformed, control charts can be established. For example, the huge amount of data is a feature of some advanced processes, and the existing statistical process control method cannot obtain an effective control boundary for monitoring in a short time.
Disclosure of Invention
The present application mainly aims to provide an alarm monitoring method and apparatus, so as to at least solve the technical problem that it takes a long time to obtain an accurate control boundary for sample data of different distribution types in the prior art.
According to an aspect of the present application, there is provided an alarm monitoring method, including: collecting sample data for alarm monitoring in the production process; determining the distribution type of the sample data; acquiring an alarm monitoring strategy corresponding to the distribution type; and performing alarm monitoring on the production process by using the alarm monitoring strategy.
Preferably, the obtaining of the alarm monitoring policy corresponding to the distribution type includes: determining a control boundary of a control chart corresponding to the distribution type; the alarm monitoring of the production process by using the determined alarm monitoring strategy comprises the following steps: and performing alarm monitoring on the production process by using the control interface which determines the control chart.
Preferably, the determining the distribution type of the sample data includes: judging whether the numerical value level of the sample data is greater than a first threshold value; if the numerical level is less than or equal to the first threshold, judging that the distribution type is a single constant type or a multi-level discrete type; if the numerical value level is larger than the first threshold value, judging whether the sample data is in normal distribution; if the sample data is normally distributed, the distribution type is judged to be a normal distribution type, and if the sample data is not normally distributed, the distribution type is judged to be a continuous non-normal distribution type, or a periodic trend upward/downward type, or a total drift type after PM maintenance is carried out regularly.
Preferably, the determining that the distribution type is a single constant type or a multi-level discrete type includes: if the value level is equal to 1, judging that the distribution type is a single constant type; otherwise, judging that the distribution type is a multi-level discrete type; the control boundary comprises an upper control boundary and a lower control boundary used for alarm monitoring, and the control boundary for determining the control chart corresponding to the distribution type comprises the following steps: setting both the upper control bound and the lower control bound to a predetermined first constant when the distribution type is the single constant type; when the distribution type is the multi-level discrete type, setting the upper control bound to a maximum value in the sample data, and setting the lower control bound to a minimum value in the sample data.
Preferably, the determining that the distribution type is a continuous non-normal distribution type, or a periodic trend up/down type, or a PM post-population drift type includes: judging whether the distribution of the sample data meets the PM change condition; if the distribution of the sample data does not meet the PM change condition, judging that the distribution type is a continuous non-normal distribution type; if the distribution of the sample data meets the PM variation condition, the distribution type is judged to be a periodic trend upward/downward type, or a PM rear total drift type.
Preferably, the determining the control boundary of the control map corresponding to the distribution type includes: when the distribution type is judged to be a continuous non-normal distribution type and the control graph is a non-normal data bilateral control graph, the control boundary comprises an upper control boundary and a lower control boundary for alarm monitoring, the upper control boundary is set to be a (100-p/2) th percentile, the lower control boundary is set to be a (p/2) th percentile, and p% is a preset alarm rate; when the distribution type is judged to be a continuous non-normal distribution type and the control chart is a non-normal data single-side control chart, the control boundary comprises an upper control boundary used for alarm monitoring, and the upper control boundary is set to be a (100-p) -th percentile, or the control boundary comprises a lower control boundary used for alarm monitoring, and the lower control boundary is set to be a (p) -th percentile.
The determining a control bound for the control chart corresponding to the distribution type includes: when the distribution type is judged to be a periodic trend upward/downward type or a PM rear total drift type, the control boundary comprises an upper control boundary for alarm monitoring, and the upper control boundary is set according to the following formula according to the difference value between every two adjacent data in the sample data:
R=|xi-xi-1|
wherein x isiThe ith data in the sample data is represented, N represents the number of the sample data, UCL represents the upper control boundary, c is an alarm rate adjustment factor, and c is obtained by solving the following formula:
Preferably, for sample data with a continuous non-normal distribution type, or a periodic trend up/down type, or a post-PM total drift type, the collecting sample data for alarm monitoring in the production process includes: at least N sample data for alarm monitoring are used in the production process, wherein,p% is the predetermined alarm rate.
Preferably, the determining the control boundary of the control map corresponding to the distribution type includes: when the distribution type is judged to be a normal distribution type and the control graph is a normal data unilateral control graph, the control boundary comprises an upper control boundary UCL or a lower control boundary LCL for alarm monitoring, and the upper control boundary UCL or the lower control boundary LCL is calculated according to a preset formula; when the distribution type is judged to be a normal distribution type and the control graph is a normal data bilateral control graph, the control boundary comprises an upper control boundary UCL and a lower control boundary LCL for alarm monitoring, and the upper control boundary UCL and the lower control boundary LCL are calculated according to the preset formula;
wherein the predetermined formula is:
UCL=u+c×S
LCL=u-c×S
wherein x isiThe ith data in the sample data is represented, N represents the number of the sample data, UCL represents the upper control boundary, LCL represents the lower control boundary, c is an alarm rate adjustment factor, and c is obtained by solving the following formula;
According to another aspect of the present application, there is provided an alarm monitoring device, comprising: the acquisition unit is used for acquiring sample data for alarm monitoring in the production process; a determining unit, configured to determine a distribution type of the sample data; the acquisition unit is used for acquiring an alarm monitoring strategy corresponding to the distribution type; and the monitoring unit is used for carrying out alarm monitoring on the production process by using the alarm monitoring strategy.
Preferably, the acquisition unit includes: the determining module is used for determining a control boundary of the control chart corresponding to the distribution type; the monitoring unit includes: and the monitoring module is used for performing alarm monitoring on the production process by using the control boundary which determines the control chart.
Through the technical scheme of this application, can reach following beneficial effect:
1) by selecting the control boundaries corresponding to different distribution types, the control boundaries for Alarm monitoring can be quickly determined for the sample data of different distribution types, the problem that the sample data of different distribution types needs to take a long time to obtain an accurate control boundary in the prior art is solved, and therefore a statistical process control method can be quickly established, so that false Alarm probability (FAR) is reduced, and abnormal fluctuation of a large amount of data is effectively monitored.
2) The system automatically judges the distribution type of the sample data and selects the control boundary corresponding to the distribution type, so that automatic systematic monitoring can be realized, and an engineer can be reminded of possible abnormal phenomena in real time.
Of course, it is not necessary for any product to achieve all of the above-described advantages at the same time for practicing the present application
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a preferred flow chart of an alarm monitoring method according to an embodiment of the present application;
FIG. 2 is another preferred flow chart of an alarm monitoring method according to an embodiment of the present application;
FIG. 3 is a preferred diagram of sample data distribution according to an embodiment of the present application; and
fig. 4 is a schematic diagram of a preferred structure of an alarm monitoring device according to an embodiment of the present application.
Detailed Description
The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
As used herein, the term "module" or "unit" may refer to software objects or routines that execute on a hardware processing device. The different modules and units described herein may be implemented as objects or processes that execute on a hardware processing device (e.g., as separate threads). Although the systems and methods described herein are preferably implemented in software, implementations in hardware or a combination of software and hardware are also possible and contemplated.
Example 1
Fig. 1 is a preferred flowchart of an alarm monitoring method according to an embodiment of the present application. As shown in fig. 1, an alarm monitoring method according to an embodiment of the present application includes:
s102, collecting sample data for alarm monitoring in the production process;
in this embodiment, the distribution types of the sample data include, but are not limited to, the following types: a single constant Type (Type 1), a multilevel discrete Type (Type 2), a continuous non-normal Type (Type 3), a normal distribution Type (Type 4), a Periodic trend up/down (Type 5) Type, and a PM (Periodic Maintenance) post-global drift Type (Type 6).
Preferably, for sample data with a distribution type of continuous non-normal distribution, or a periodic trend upward/downward type, or a post-PM total drift type, the collecting sample data for alarm monitoring in the production process includes: adopting at least N sample data for alarm monitoring in the production process, wherein,p% is the predetermined alarm rate.
S104, determining the distribution type of the sample data;
preferably, the determining the distribution type of the sample data includes: judging whether the numerical value level of the sample data is greater than a first threshold value; if the numerical value level is less than or equal to the first threshold, judging that the distribution type is a single constant type or a multi-level discrete type; if the numerical value level is larger than the first threshold value, judging whether the sample data is in normal distribution; and if the sample data is normally distributed, judging that the distribution type is a normal distribution type, and if the sample data is not normally distributed, judging that the distribution type is a continuous non-normal distribution type, or a periodic trend upward/downward type, or a PM rear total drift type.
S106, acquiring an alarm monitoring strategy corresponding to the distribution type;
preferably, the obtaining of the alarm monitoring policy corresponding to the distribution type includes: and determining the control boundary of the control chart corresponding to the distribution type. If the control graph is a bilateral control graph, the control boundary comprises an upper control boundary and a lower control boundary which are used for alarm monitoring; if the control chart is a single-side control chart, the control boundary comprises an upper control boundary or a lower control boundary for alarm monitoring.
The following describes a preferred process for determining a control boundary of a control map in the present embodiment, where UCL represents an upper control boundary and LCL represents a lower control boundary:
1) single constant type: UCL ═ LCL ═ Target ═ predetermined constant;
2) multi-level discrete type: UCL ═ maximum constant value (e.g., maximum value in the sample data), LCL ═ minimum constant value (e.g., minimum value in the sample data), Target ═ mean value of the sample data;
3) continuous non-normal distribution: target = mean value of sample data, when the control chart is a non-normal data bilateral control chart, UCL = the (100-p/2) th percentile, LCL = the (p/2) th percentile, where p% is a predetermined alarm rate; when the control chart is a non-normal data one-sided control chart, UCL = the (100-p) th percentile, or LCL = the (p) th percentile;
the percentile described above is a statistical term, and the (p) th percentile indicates that, among a set of sample data sorted from small to large, the number of sample data smaller than or equal to UCL accounts for p% of the total number of sample data, and the number of sample data larger than UCL accounts for 1-p% of the total number of sample data. For example, the total number of sample data collected is 200, the value of UCL is 50, which corresponds to the 70 th percentile, which means: in the 200 collected sample data, 70% of the sample data is less than or equal to 50, and 30% of the sample data is greater than 50.
4) Normal distribution type: target = mean value of sample data u, UCL = u + c × S, LCL = u-c × S;
UCL=u+c×S
LCL=u-c×S
wherein x isiAnd representing ith data in the sample data, N representing the number of the sample data, UCL representing the upper control boundary, LCL representing the lower control boundary, c being an alarm rate adjustment factor, and c being obtained by solving the following equation.
5) -6) PM periodic tend-down/tend-up type and PM post-gross drift type: the control boundary comprises an upper control boundary used for alarm monitoring, and the upper control boundary is set according to the following formula according to the difference value between every two adjacent data in the collected sample data:
R=|xi-xi-1|
wherein x isiRepresenting ith data in the sample data, N representing the number of the sample data, UCL representing the upper boundary, c being an alarm rate adjustment factor, c being obtained by solving the following equation:
And S108, performing alarm monitoring on the production process by using the determined alarm monitoring strategy.
Preferably, the performing alarm monitoring on the production process by using the determined alarm monitoring policy includes: and performing alarm monitoring on the production process by using the control boundary determining the control chart.
In the embodiment, the control boundary corresponding to different distribution types is selected, so that the control boundary for alarm monitoring can be quickly determined for the sample data of different distribution types, and the technical problem that the sample data of different distribution types needs to take a long time to obtain an accurate control boundary in the prior art is solved, so that a statistical process control method can be quickly established, the false alarm probability is reduced, and the abnormal fluctuation of a large amount of data is effectively monitored.
The alarm monitoring method according to the embodiment of the present application will be described in detail with reference to fig. 2.
As shown in fig. 2, the flow of automatic generation of the quality control chart includes the following steps:
step 201, data acquisition. And automatically starting to generate the control diagram after the acquired data exceeds n points. The value of n can be obtained by the following formula according to the false alarm rate required by engineering.
Wherein,
wherein α is the confidence, typically 0.01,0.05, or 0.1.;
represents; cumulative probability of t distribution with degree of freedom (n-1) ofA value of (d);
represents: the cumulative probability of chi-square distribution with the degree of freedom (n-1) isA value of (d);
s represents: standard deviation of sample data;
represents: a mean value of the sample data;
n represents: the number of sample data.
Step 202, determine whether the value level of the data is less than or equal to the first threshold (in this embodiment, the value level of the data represents the number of values that the data may be equal to, for example, if sample data a =2 or 3, the value level of the sample data a is 2). For example, in this embodiment, the first threshold is 10, and when the numerical level of the data < =10, step 203 is performed to perform discrete distribution determination and monitoring; when the value level of the data is >10, the process proceeds to step 204 for continuous profile monitoring. The first threshold is 10, which is only an example, and the present embodiment is not limited thereto, and the first threshold may be adjusted according to the number of the collected sample data, for example, when the number of the collected sample data is about 100, the first threshold may be 10.
Step 203, judging whether the numerical value level of the sample data is a second threshold value. In this embodiment, the second threshold =1, but the present embodiment is not limited thereto, and the second threshold may be adjusted according to the number of the collected sample data.
When the data level of the sample data =1, judging that the distribution Type of the sample data is a single constant Type (Type 1), and monitoring by adopting a control boundary of the following control chart: UCL ═ LCL ═ Target ═ a predetermined constant.
When the data level of the sample data is not equal to 1, judging that the distribution Type of the sample data is a multi-level discrete Type (Type 2), and monitoring by adopting a control boundary of the following control chart: UCL is the maximum constant value (e.g., the maximum value in the sample data), LCL is the minimum constant value (e.g., the minimum value in the sample data), and Target is the mean of the sample data.
And 204, judging whether the distribution type of the sample data is a normal distribution type. In this embodiment, an existing test scheme may be used to determine whether the sample data is normally distributed, and different test schemes may be used according to the requirements of different situations, for example, a Shipro-Wilk normal distribution test scheme, a D' agonstino test scheme, a Kolmogorov AN. test scheme, and the like may be used. For example, if the P value of the test result is greater than 0.05, the distribution type of the sample data is determined to be a normal distribution type, and the process goes to step 205 to perform normal distribution monitoring; and if the P value of the test result is less than or equal to 0.05, judging that the distribution type of the sample data is not a normal distribution type, and turning to step 206 to perform non-normal distribution control, wherein the P value of the test result represents the probability of misjudgment, and the misjudgment means that the distribution of the sample data is not a normal distribution as the result of the judgment under the condition that the distribution of the sample data is actually a normal distribution. The verification scheme described above can be derived from programming and existing statistical tool modules.
Step 205, judging that the distribution Type of the sample data is a normal distribution Type (Type 4), which can be, but is not limited to, monitoring by adopting the following steps:
step S1 removes outliers. Before the control boundary calculation, outliers in the sample data are removed, so that the estimation precision is improved. In this embodiment, whether the sample data is an outlier may be determined in the following manner: judging whether the sample data is located in a range (Q1-x IQR, Q1+ x IQR), wherein Q1= 25 th percentile, IQR = 75 th percentile-25 th percentile, and x is 1.5 and 3 …; if the sample data is out of the range, the sample data is judged to be an outlier. Of course, the above-mentioned manner of removing outliers is only an example, and the present application does not limit this.
Step S2 determines whether the normal data single-sided control chart is used. After outliers are removed, it is automatically determined whether to use normal data one-sided control graph (expected big data or expected small data).
Step S2.1 determines whether one of the following two conditions is met: 1) only one side has SPEC limit (Upper or lower SPEC limit, USL or LSL for short); 2) target equals SPEC on one side.
And S2.2, if one of the two conditions is met, judging that the normal data single-sided control chart is adopted, otherwise, judging that the normal data double-sided control chart is adopted.
And S2.3, when judging that the normal data unilateral control graph is adopted, further judging as follows: if only USL or Target = LSL, the control method is an upper limit single-side control chart; on the contrary, if only LSL or Target = USL, the lower limit single-sided control map is obtained.
Step S3 calculates a control limit.
Step S3.1 specifically, for the normal-data bilateral control map, Target = mean u of the sample data, UCL = u + c × S, LCL = u-c × S, UCA and LCL are used as the upper control limit/lower control limit, respectively.
UCL=u+c×S
LCL=u-c×S
Wherein x isiRepresenting the ith data in the sample data, N representing the number of the sample data, UCL representing the upper control boundary,LCL represents the lower control bound, c is an alarm rate adjustment factor, and c can be solved by the following equation.
Step S3.2 uses Target = mean u of sample data, UCL = u + c × S, LCL = u-c × S, UCA and LCL as upper/lower control boundary, respectively, for the normal data single-sided control map.
UCL=u+c×S
or
LCL=u-c×S
Wherein x isiRepresenting ith data in the sample data, N representing the number of the sample data, UCL representing the upper control boundary, LCL representing the lower control boundary, c being an alarm rate adjustment factor, and c being obtained by solving the following equation:
And step 206, judging whether the distribution of the sample data meets the PM change condition.
Due to Periodic Maintenance (PM) in the production process, the distribution of the collected sample data may meet a predetermined PM change condition. In this embodiment, the collected sample data includes sample data in a plurality of periods, and the PM variation condition may include, but is not limited to: the difference between the mean values of the sample data in two adjacent periods is greater than a predetermined threshold, for example, the mean value of the sample data acquired in the first period is 100, and the mean value of the sample data acquired in the second period is 200. If the difference between the mean values of sample data in two adjacent periods is greater than a predetermined threshold (for example, 50), the distribution of the collected sample data is considered to satisfy a predetermined PM change condition, or a PM Shift (PM Shift) is considered to occur.
In this embodiment, the above-mentioned Periodic Maintenance (PM) may include, but is not limited to: the chemical solvent is replaced for the chemical etching process. Thus, the concentration of the chemical solvent is increased due to the replacement of the chemical solvent, so that the etching rate is suddenly increased, and then the concentration of the chemical solvent is gradually decreased with the continuous use of the chemical solvent, so that the etching rate is reduced with the time. The periodic maintenance described above causes the mean value of the sample data collected in two periods to change. The period mentioned above may be a time interval between two adjacent sample data acquisitions, for example, when the period =1 hour, it means that sample data is acquired every 1 hour.
If the distribution of the sample data does not meet the PM change condition, judging that the distribution Type of the sample data is a continuous non-normal distribution Type (Type 3); otherwise, judging that the distribution type of the sample data is a periodic trend upward/downward type or a PM rear total drift type.
Step 207, judging that the distribution Type of the sample data is a continuous non-normal distribution Type (Type 3), and monitoring by adopting the following steps:
step S1 removes outliers. For example, outliers are removed by removing the oos (out of spec) value, which may be a predetermined threshold.
Step S2 determines whether to use the non-normal data single-sided control chart or the non-normal data double-sided control chart, and the specific determination process may refer to S2 in step S205.
Step S3 calculates a control limit.
Step S3.1 uses the (100-p/2) th percentile and the (p/2) th percentile as an upper control boundary and a lower control boundary for the non-normal data bilateral control chart, wherein p% is a preset alarm rate.
And S3.2, for the non-normal data unilateral control chart, using the (100-p) th percentile or the (p) th percentile as an upper control boundary or a lower control boundary, and respectively using the upper limit unilateral control chart or the lower limit unilateral control chart.
And 208, judging that the distribution Type of the sample data is a periodic trend-up/trend-down Type (Type 5) or a PM post-total drift Type (Type 6). According to the setting carried out by the background of the system, after the distribution Type of the sample data is judged to be a periodic trand-up/trand-down Type (Type 5) or a PM post-total drift Type (Type 6), an MR (MoveRange) control chart is automatically established, namely, an MR (MoveRange) control chart is automatically established according to the difference (x) of the two measurement values before and after the sample data passesi-xi-1) And (5) monitoring. Preferably, the control bound can be calculated by the following formula.
R=|xi-xi-1|
Wherein x isiRepresenting ith data in the sample data, N representing the number of the sample data, UCL representing the upper boundary, c being an alarm rate adjustment factor, c being obtained by solving the following equation:
In the above embodiment, the number of the collected sample data and the control of the false alarm rate may be selected based on the following scheme:
for a non-normal distribution, since the most edge needs to estimate the (p/2) th percentile, the number of the collected sample data needs to estimate the reciprocal of the percentile at least, namely (2/p%) data. For example, if the predetermined alarm rate is 1%, then the minimum percentile estimated for the non-normal bilateral control chart is 0.5%, so at least 200 points of data are needed.
For normal distribution, because the sample is used to estimate the overall parameters, the estimated error will be different according to the number of sample data, the number of sample data is defined as n, and the mean value of the sample data is defined as nThe variance of the sample data is S2, the confidence interval of the mean and variance estimates of the sample data is the following equation [2]]And [3]。
Confidence interval for μ: formula [2]
σ2Confidence interval of (c): formula [3]
[(n-1)s2/χ1-α/2 2(n-1),(n-1)s2/χα/2 2(n-1)]
Wherein,
α is a confidence, typically 0.01,0.05, or 0.1.;
represents: the chi-square distribution cumulative probability with the degree of freedom (n-1) is a value of 1-alpha/2;
t1-α/2(n-1); the cumulative probability of t distribution with degree of freedom (n-1) is 1-alpha/2;
s represents: standard deviation of sample data;
represents: a mean value of the sample data;
n represents: the number of sample data.
From the above formula, it can be seen that there are cases where the estimated mean deviates from the overall mean and the estimated variance is smaller than the overall variance, as shown in fig. 3. The mean and variance estimates after normalization are respectivelyAndtherefore, using the (1- α) confidence interval estimate (typically α is chosen to be 0.05), taking into account that the maximum value is reached in the estimated meanThe standard deviation of the estimation reaches the minimum valueAnd the alarm rate is maximum. The false alarm rate p is therefore a function of the number of sample data, the functional formula being [2]]And [3]From this, the number n of the smallest sample data can be calculated with the known false alarm rate p, the confidence α (Alpha), and the number of WECO rules that need to be used.
Table (b): false alarm rate estimated for different sample data numbers (e.g. Alpha =0.05,2WECO rule, minimum sample number n is 200)
For example, in practical use, 2WECO rule samples are selected, alpha is 0.05, the false alarm rate is 1%, and the minimum sample number is 200 by table lookup.
Therefore, in order to simultaneously achieve a false alarm rate of less than 1% under the normal distribution control chart and the non-normal distribution control chart, 200 sample data are acquired at minimum. If the number of the minimum sample data calculated by the method is different from the number of the minimum sample data obtained by table lookup, the larger number of the minimum sample data and the minimum sample data is taken as the number of the minimum sample data.
On the basis of the solutions shown in fig. 1 to fig. 3, the present application further provides an alarm monitoring device, as shown in fig. 4, the alarm monitoring device includes:
1) an acquisition unit 402, configured to acquire sample data for alarm monitoring in a production process;
in this embodiment, the distribution types of the sample data include, but are not limited to, the following types: a single constant Type (Type 1), a multilevel discrete Type (Type 2), a continuous non-normal Type (Type 3), a normal distribution Type (Type 4), a periodic trend up/down (Type 5) Type, and a PM post-total drift Type (Type 6).
Preferably, for sample data with a distribution type of continuous non-normal distribution, or a periodic trend upward/downward type, or a post-PM total drift type, the collecting sample data for alarm monitoring in the production process includes: adopting at least N sample data for alarm monitoring in the production process, wherein,p% is the predetermined alarm rate.
2) A determining unit 404, configured to determine a distribution type of the sample data;
preferably, the determining the distribution type of the sample data includes: judging whether the numerical value level of the sample data is greater than a first threshold value; if the numerical value level is less than or equal to the first threshold, judging that the distribution type is a single constant type or a multi-level discrete type; if the numerical value level is larger than the first threshold value, judging whether the sample data is in normal distribution; and if the sample data is normally distributed, judging that the distribution type is a normal distribution type, and if the sample data is not normally distributed, judging that the distribution type is a continuous non-normal distribution type, or a periodic trend upward/downward type, or a PM rear total drift type.
Preferably, the distribution type of the sample data may be determined by the method described above (as shown in fig. 2), which is not described herein again.
3) An obtaining unit 406, configured to obtain an alarm monitoring policy corresponding to the distribution type;
preferably, the obtaining of the alarm monitoring policy corresponding to the distribution type includes: and determining the control boundary of the control chart corresponding to the distribution type. If the control graph is a bilateral control graph, the control boundary comprises an upper control boundary and a lower control boundary which are used for alarm monitoring; if the control chart is a single-side control chart, the control boundary comprises an upper control boundary or a lower control boundary for alarm monitoring.
The following describes a preferred process for determining a control boundary of a control map in the present embodiment, where UCL represents an upper control boundary and LCL represents a lower control boundary:
1) single constant type: UCL ═ LCL ═ Target ═ predetermined constant;
2) multi-level discrete type: UCL ═ maximum constant value (e.g., maximum value in sample data), LCL ═ minimum constant value (e.g., minimum value in sample data), Target (Target) is the mean of sample data;
3) continuous non-normal distribution: target = mean value of sample data, when the profile is a non-normal data bilateral profile, UCL = the (100-p/2) th percentile, LCL = the (p/2) th percentile, where p% is a predetermined alarm rate; when the control chart is a non-normal data one-sided control chart, UCL = the (100-p) th percentile, or LCL = the (p) th percentile;
4) normal distribution type: target = mean value u of sample data, UCL = u + c × S, LCL = u-c × S,
UCL=u+c×S
LCL=u-c×S
wherein x isiRepresenting ith data in the sample data, N representing the number of the sample data, UCL representing the upper control boundary, LCL representing the lower control boundary, c being an alarm rate adjustment factor, and c being obtained by solving the following equation:
5) -6) PM periodic trend up/down (trand-down/trand-up) type and PM post-global drift type: the control boundary comprises an upper control boundary used for alarm monitoring, and the upper control boundary is set according to the following formula according to the difference value between every two adjacent data in the collected sample data:
R=|xi-xi-1|
wherein x isiRepresenting ith data in the sample data, N representing the number of the sample data, UCL representing the upper boundary, c being an alarm rate adjustment factor, c being obtained by solving the following equation:
Preferably, the control boundary of the control map can be determined in the manner described above (as shown in fig. 2), and the description of the present application is omitted here.
4) And a monitoring unit 408, configured to perform alarm monitoring on the production process by using the alarm monitoring policy.
Preferably, the obtaining unit 406 includes: and the determining module is used for determining the control boundary of the control chart corresponding to the distribution type. The monitoring unit 408 includes: and the monitoring module is used for performing alarm monitoring on the production process by using the control boundary which determines the control chart.
In the embodiment, the control boundary corresponding to different distribution types is selected, so that the control boundary for alarm monitoring can be quickly determined for the sample data of different distribution types, and the technical problem that the sample data of different distribution types needs to take a long time to obtain an accurate control boundary in the prior art is solved, so that a statistical process control method can be quickly established, the false alarm probability is reduced, and the abnormal fluctuation of a large amount of data is effectively monitored.
It will be apparent to those skilled in the art that the modules or steps of the present application described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and alternatively, they may be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, the present application is not limited to any specific combination of hardware and software.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (11)
1. An alarm monitoring method, comprising:
collecting sample data for alarm monitoring in the production process;
determining the distribution type of the sample data;
acquiring an alarm monitoring strategy corresponding to the distribution type;
and performing alarm monitoring on the production process by using the alarm monitoring strategy.
2. The method of claim 1,
the acquiring of the alarm monitoring strategy corresponding to the distribution type includes: determining a control boundary of a control chart corresponding to the distribution type;
the alarm monitoring of the production process by using the determined alarm monitoring strategy comprises the following steps: and performing alarm monitoring on the production process by using the control boundary determining the control chart.
3. The method of claim 2, wherein said determining the type of distribution of the sample data comprises:
judging whether the numerical value level of the sample data is greater than a first threshold value;
if the numerical value level is less than or equal to the first threshold, judging that the distribution type is a single constant type or a multi-level discrete type;
if the numerical value level is larger than the first threshold value, judging whether the sample data is in normal distribution; and if the sample data is normally distributed, judging that the distribution type is a normal distribution type, and if the sample data is not normally distributed, judging that the distribution type is a continuous non-normal distribution type, or a periodic trend upward/downward type, or a total drift type after PM maintenance is performed regularly.
4. The method of claim 3,
the judging that the distribution type is a single constant type or a multi-level discrete type includes: if the numerical level is equal to 1, judging that the distribution type is a single constant type; otherwise, judging that the distribution type is a multi-level discrete type;
the control boundary comprises an upper control boundary and a lower control boundary used for alarm monitoring, and the determining of the control boundary of the control chart corresponding to the distribution type comprises the following steps: when the distribution type is the single constant type, setting the upper control boundary and the lower control boundary as a preset first constant; when the distribution type is the multi-level discrete type, setting the upper control bound as a maximum value in the sample data, and setting the lower control bound as a minimum value in the sample data.
5. The method of claim 3, wherein determining the distribution type is a continuous non-normal distribution type, or a periodic trend up/down type, or a PM post population drift type comprises:
judging whether the distribution of the sample data meets a PM change condition;
if the distribution of the sample data does not meet the PM change condition, judging that the distribution type is a continuous non-normal distribution type;
and if the distribution of the sample data meets the PM change condition, judging that the distribution type is a periodic trend upward/downward type or a PM rear total drift type.
6. The method of claim 5, wherein determining the control bound for the control chart corresponding to the distribution type comprises:
when the distribution type is judged to be a continuous non-normal distribution type and the control graph is a non-normal data bilateral control graph, the control boundary comprises an upper control boundary and a lower control boundary for alarm monitoring, the upper control boundary is set to be a (100-p/2) th percentile, the lower control boundary is set to be a (p/2) th percentile, and p% is a preset alarm rate;
and when the distribution type is judged to be a continuous non-normal distribution type and the control chart is a non-normal data unilateral control chart, setting the upper control boundary to be a (100-p) -th percentile by using the control boundary as an upper control boundary for alarm monitoring, or setting the lower control boundary to be a (p) -th percentile by using the control boundary as a lower control boundary for alarm monitoring.
7. The method of claim 5, wherein determining the control bound for the control chart corresponding to the distribution type comprises:
when the distribution type is judged to be a periodic trend upward/downward type or a PM rear total drift type, the control boundary comprises an upper control boundary for alarm monitoring, and the upper control boundary is set according to the following formula according to the difference value between every two adjacent data in the sample data:
R=|xi-xi-1|
wherein x isiRepresenting ith data in the sample data, N representing the number of the sample data, UCL representing the upper control bound, c being an alarm rate adjustment factor, c being solved by the following formula:
8. The method according to claim 3, wherein for sample data with the distribution type of continuous non-normal distribution type, or periodic trend up/down type, or post-PM total drift type, the collecting sample data for alarm monitoring in the production process comprises: adopting at least N sample data for alarm monitoring in the production process, wherein,p% is the predetermined alarm rate.
9. The method of claim 5, wherein determining the control bound for the control chart corresponding to the distribution type comprises:
when the distribution type is judged to be a normal distribution type and the control graph is a normal data unilateral control graph, the control boundary comprises an upper control boundary UCL or a lower control boundary LCL for alarm monitoring, and the upper control boundary UCL or the lower control boundary LCL is calculated according to a preset formula;
when the distribution type is judged to be a normal distribution type and the control graph is judged to be a normal data bilateral control graph, the control boundary comprises an upper control boundary UCL and a lower control boundary LCL which are used for alarm monitoring, and the upper control boundary UCL and the lower control boundary LCL are calculated according to the preset formula;
wherein the predetermined formula is:
UCL=u+c×S
LCL=u-c×S
wherein x isiRepresenting ith data in the sample data, N representing the number of the sample data, UCL representing the upper control boundary, LCL representing the lower control boundary, c being an alarm rate adjustment factor, and c being obtained by solving the following formula;
10. An alarm monitoring device, comprising:
the acquisition unit is used for acquiring sample data for alarm monitoring in the production process;
a determining unit, configured to determine a distribution type of the sample data;
the acquisition unit is used for acquiring the alarm monitoring strategy corresponding to the distribution type;
and the monitoring unit is used for carrying out alarm monitoring on the production process by using the alarm monitoring strategy.
11. The apparatus of claim 10,
the acquisition unit includes: the determining module is used for determining a control boundary of the control chart corresponding to the distribution type;
the monitoring unit includes: and the monitoring module is used for performing alarm monitoring on the production process by using the control boundary which determines the control chart.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310066956.7A CN104020724B (en) | 2013-03-01 | 2013-03-01 | alarm monitoring method and device |
US14/017,297 US9798320B2 (en) | 2013-03-01 | 2013-09-03 | Method and apparatus for alarm monitoring |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310066956.7A CN104020724B (en) | 2013-03-01 | 2013-03-01 | alarm monitoring method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104020724A true CN104020724A (en) | 2014-09-03 |
CN104020724B CN104020724B (en) | 2017-02-08 |
Family
ID=51421351
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310066956.7A Active CN104020724B (en) | 2013-03-01 | 2013-03-01 | alarm monitoring method and device |
Country Status (2)
Country | Link |
---|---|
US (1) | US9798320B2 (en) |
CN (1) | CN104020724B (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105302123A (en) * | 2015-11-25 | 2016-02-03 | 上海大众汽车有限公司 | Online data monitoring method |
CN105718742A (en) * | 2016-01-22 | 2016-06-29 | 张燕 | Dynamic hidden danger early warning method based on statistics |
CN105988459A (en) * | 2015-02-11 | 2016-10-05 | 中芯国际集成电路制造(上海)有限公司 | Method for predicting machine fault based on mean small drift |
CN106649569A (en) * | 2016-11-16 | 2017-05-10 | 上海建桥学院 | Data analysis and early-warning method based on information connotation degree |
CN106960554A (en) * | 2017-03-20 | 2017-07-18 | 山东科技大学 | Warning system evaluation method and device based on single simulation amount and presignal delay device |
CN107093568A (en) * | 2016-02-18 | 2017-08-25 | 北大方正集团有限公司 | A kind of wafer on-line monitoring method and device |
CN108150361A (en) * | 2016-12-05 | 2018-06-12 | 北京金风科创风电设备有限公司 | The monitoring method and device of wind generating set engine room vibration |
CN108270618A (en) * | 2017-12-30 | 2018-07-10 | 杭州华为数字技术有限公司 | Alert the method, apparatus and warning system of judgement |
CN108628281A (en) * | 2017-03-23 | 2018-10-09 | 株式会社日立制作所 | Abnormality detection system and method for detecting abnormality |
CN109408303A (en) * | 2018-09-19 | 2019-03-01 | Oppo(重庆)智能科技有限公司 | Test data analysis method and Related product |
CN109873853A (en) * | 2017-12-05 | 2019-06-11 | 北京绪水互联科技有限公司 | Equipment key parameter early warning system and its implementation, electronic device |
CN110050242A (en) * | 2017-01-26 | 2019-07-23 | 三菱日立电力系统株式会社 | Warning display system and alarm display methods |
CN111626841A (en) * | 2020-04-10 | 2020-09-04 | 中国建设银行股份有限公司 | Method, system and related equipment for monitoring online transaction |
CN111855178A (en) * | 2020-07-23 | 2020-10-30 | 贵州永红航空机械有限责任公司 | Diagnosis method for running state of rotary product |
CN113626906A (en) * | 2021-07-07 | 2021-11-09 | 武汉轻工大学 | Underground supporting structure deformation monitoring method based on data mining and control chart |
CN115913896A (en) * | 2022-11-09 | 2023-04-04 | 中国联合网络通信集团有限公司 | Device detection method, server and medium |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190195695A1 (en) * | 2017-08-30 | 2019-06-27 | Walmart Apollo, Llc | Temperature environmental sensors on pallet |
CN109374053B (en) * | 2018-11-13 | 2021-06-08 | 深圳市中广控信息科技有限公司 | Internet of things machine room management platform based on event-driven response |
CN111327442B (en) * | 2018-12-17 | 2022-08-30 | 中国移动通信集团广东有限公司 | Complaint early warning threshold value obtaining method and device based on control chart |
CN109685799B (en) * | 2018-12-28 | 2020-11-13 | 北京航空航天大学 | Statistical process monitoring method based on image data and false alarm rate of given interval |
CN111752251A (en) * | 2020-07-07 | 2020-10-09 | 临沂启阳电缆有限公司 | Intelligent control management system for factory workshop |
DE102022117924A1 (en) * | 2022-07-18 | 2024-01-18 | Infineon Technologies Ag | METHOD AND DEVICE FOR MONITORING A PROCESS |
CN117457125A (en) * | 2023-11-15 | 2024-01-26 | 中航通飞华南飞机工业有限公司 | Method for evaluating and analyzing performance stability of small-batch civil aircraft materials |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1058172A2 (en) * | 1999-06-01 | 2000-12-06 | Applied Materials, Inc. | Semiconductor processing techniques |
CN101290517A (en) * | 2007-04-17 | 2008-10-22 | 中芯国际集成电路制造(上海)有限公司 | Method and device for statistical process control for discrete samples data |
CN101477358A (en) * | 2008-01-04 | 2009-07-08 | 中芯国际集成电路制造(上海)有限公司 | Statistical process control method and apparatus |
EP2333630A2 (en) * | 2008-01-07 | 2011-06-15 | Metryx Limited | Method of Controlling Semiconductor Device Fabrication |
CN102540944A (en) * | 2012-01-13 | 2012-07-04 | 顺德职业技术学院 | Embedded multifunctional statistical process control (SPC) device and method |
Family Cites Families (41)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5257206A (en) * | 1991-04-08 | 1993-10-26 | Praxair Technology, Inc. | Statistical process control for air separation process |
US5555191A (en) * | 1994-10-12 | 1996-09-10 | Trustees Of Columbia University In The City Of New York | Automated statistical tracker |
SE513188C2 (en) * | 1998-03-25 | 2000-07-24 | Ericsson Telefon Ab L M | Quota algorithm for monitoring disturbance processes |
US5987398A (en) * | 1998-04-30 | 1999-11-16 | Sony Corporation | Method and apparatus for statistical process control of machines and processes having non-constant mean of a response variable |
US7073198B1 (en) * | 1999-08-26 | 2006-07-04 | Ncircle Network Security, Inc. | Method and system for detecting a vulnerability in a network |
US6728658B1 (en) * | 2001-05-24 | 2004-04-27 | Simmonds Precision Products, Inc. | Method and apparatus for determining the health of a component using condition indicators |
US7136794B1 (en) * | 2001-05-24 | 2006-11-14 | Simmonds Precision Products, Inc. | Method and apparatus for estimating values for condition indicators |
US6643613B2 (en) * | 2001-07-03 | 2003-11-04 | Altaworks Corporation | System and method for monitoring performance metrics |
US7096074B2 (en) * | 2002-05-30 | 2006-08-22 | Insyst Ltd. | Methods and apparatus for early fault detection and alert generation in a process |
US6820022B2 (en) * | 2002-08-07 | 2004-11-16 | Kimberly-Clark Worldwide, Inc. | System and method for identifying and exploiting quality information associated with a web converting manufacturing process |
US7123981B2 (en) * | 2002-08-07 | 2006-10-17 | Kimberly-Clark Worldwide, Inc | Autosetpoint registration control system and method associated with a web converting manufacturing process |
US7383071B1 (en) * | 2003-04-25 | 2008-06-03 | United States Of America As Represented By The Secretary Of The Navy | Microsensor system and method for measuring data |
US7451003B2 (en) * | 2004-03-04 | 2008-11-11 | Falconeer Technologies Llc | Method and system of monitoring, sensor validation and predictive fault analysis |
US7260501B2 (en) * | 2004-04-21 | 2007-08-21 | University Of Connecticut | Intelligent model-based diagnostics for system monitoring, diagnosis and maintenance |
EP1766484B1 (en) * | 2004-06-12 | 2013-08-07 | Fisher-Rosemount Systems, Inc. | System and method for detecting an abnormal situation associated with a process gain of a control loop |
KR100598398B1 (en) * | 2004-10-22 | 2006-07-07 | 삼성전자주식회사 | Compensating method of sensor data and evaluating interlock of interlock system |
JP2006277523A (en) * | 2005-03-30 | 2006-10-12 | Nec Electronics Corp | Abnormality detection system and abnormality detecting method |
WO2007087728A1 (en) * | 2006-02-03 | 2007-08-09 | Recherche 2000 Inc. | Adaptive method and system of monitoring signals for detecting anomalies |
CN100517136C (en) * | 2006-03-28 | 2009-07-22 | 中芯国际集成电路制造(上海)有限公司 | Method for manufacturing seniconductor device |
US8145358B2 (en) * | 2006-07-25 | 2012-03-27 | Fisher-Rosemount Systems, Inc. | Method and system for detecting abnormal operation of a level regulatory control loop |
US7912676B2 (en) * | 2006-07-25 | 2011-03-22 | Fisher-Rosemount Systems, Inc. | Method and system for detecting abnormal operation in a process plant |
US8606544B2 (en) * | 2006-07-25 | 2013-12-10 | Fisher-Rosemount Systems, Inc. | Methods and systems for detecting deviation of a process variable from expected values |
US7657399B2 (en) * | 2006-07-25 | 2010-02-02 | Fisher-Rosemount Systems, Inc. | Methods and systems for detecting deviation of a process variable from expected values |
WO2008025102A1 (en) * | 2006-09-01 | 2008-03-06 | Commonwealth Scientific And Industrial Research Organisation | Method and apparatus for signal recovery |
US8010292B2 (en) * | 2006-09-28 | 2011-08-30 | Fisher-Rosemount Systems, Inc. | Method and system for detecting abnormal operation in a hydrocracker |
US20080188972A1 (en) * | 2006-10-11 | 2008-08-07 | Fisher-Rosemount Systems, Inc. | Method and System for Detecting Faults in a Process Plant |
US7663828B2 (en) * | 2006-10-23 | 2010-02-16 | Hitachi Global Storage Technologies Netherlands B.V. | Determining the reliability of a disk drive |
US8032340B2 (en) * | 2007-01-04 | 2011-10-04 | Fisher-Rosemount Systems, Inc. | Method and system for modeling a process variable in a process plant |
US8032341B2 (en) * | 2007-01-04 | 2011-10-04 | Fisher-Rosemount Systems, Inc. | Modeling a process using a composite model comprising a plurality of regression models |
US8285513B2 (en) * | 2007-02-27 | 2012-10-09 | Exxonmobil Research And Engineering Company | Method and system of using inferential measurements for abnormal event detection in continuous industrial processes |
AT10073U9 (en) * | 2008-01-14 | 2009-02-15 | Avl List Gmbh | METHOD AND DEVICE FOR ANALYZING AND EVALUATING MEASUREMENT DATA OF A MEASURING SYSTEM |
US9606520B2 (en) * | 2009-06-22 | 2017-03-28 | Johnson Controls Technology Company | Automated fault detection and diagnostics in a building management system |
TWI412906B (en) * | 2010-04-13 | 2013-10-21 | Univ Nat Cheng Kung | Manufacturing execution system with virtual-metrology capabilities and manufacturing system including the same |
US20120123981A1 (en) * | 2010-08-04 | 2012-05-17 | Graves Spencer B | Software to facilitate design, data flow management, data analysis and decision support in structural health monitoring systems |
WO2012035547A2 (en) * | 2010-09-13 | 2012-03-22 | Manufacturing System Insights (India) Pvt. Ltd. | Apparatus that analyses attributes of diverse machine types and technically upgrades performance by applying operational intelligence and the process therefor |
US9245226B2 (en) * | 2010-09-27 | 2016-01-26 | Carmel-Haifa University Economic Corporation Ltd | Detecting change points in data streams |
US9952565B2 (en) * | 2010-11-15 | 2018-04-24 | Guang Liu | Networked, channelized power distribution, monitor and control for security and life safety applications |
GB2491564A (en) * | 2011-05-24 | 2012-12-12 | Isis Innovation | Method of system monitoring |
US20130041625A1 (en) * | 2011-08-11 | 2013-02-14 | International Business Machines Corporation | Advanced Statistical Detection of Emerging Trends |
GB201215649D0 (en) * | 2012-09-03 | 2012-10-17 | Isis Innovation | System monitoring |
US9152610B2 (en) * | 2012-11-16 | 2015-10-06 | Johnson Controls Technology Company | Systems and methods for generating an energy use model for a building |
-
2013
- 2013-03-01 CN CN201310066956.7A patent/CN104020724B/en active Active
- 2013-09-03 US US14/017,297 patent/US9798320B2/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1058172A2 (en) * | 1999-06-01 | 2000-12-06 | Applied Materials, Inc. | Semiconductor processing techniques |
CN101290517A (en) * | 2007-04-17 | 2008-10-22 | 中芯国际集成电路制造(上海)有限公司 | Method and device for statistical process control for discrete samples data |
CN101477358A (en) * | 2008-01-04 | 2009-07-08 | 中芯国际集成电路制造(上海)有限公司 | Statistical process control method and apparatus |
EP2333630A2 (en) * | 2008-01-07 | 2011-06-15 | Metryx Limited | Method of Controlling Semiconductor Device Fabrication |
CN102540944A (en) * | 2012-01-13 | 2012-07-04 | 顺德职业技术学院 | Embedded multifunctional statistical process control (SPC) device and method |
Non-Patent Citations (4)
Title |
---|
于信明等: "SPC在半导体生产中的应用", 《制造工艺技术》 * |
王文君: "采用SPC技术控制半导体器件的工艺质量", 《微电子器件与技术》 * |
赵会珍: "面向制造过程的质量控制系统研究及应用", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
陈建国等: "SPC技术及应用", 《微电子技术》 * |
Cited By (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105988459A (en) * | 2015-02-11 | 2016-10-05 | 中芯国际集成电路制造(上海)有限公司 | Method for predicting machine fault based on mean small drift |
CN105988459B (en) * | 2015-02-11 | 2019-01-18 | 中芯国际集成电路制造(上海)有限公司 | Method based on the small drift forecasting board failure of mean value |
CN105302123A (en) * | 2015-11-25 | 2016-02-03 | 上海大众汽车有限公司 | Online data monitoring method |
CN105718742A (en) * | 2016-01-22 | 2016-06-29 | 张燕 | Dynamic hidden danger early warning method based on statistics |
CN107093568A (en) * | 2016-02-18 | 2017-08-25 | 北大方正集团有限公司 | A kind of wafer on-line monitoring method and device |
CN106649569A (en) * | 2016-11-16 | 2017-05-10 | 上海建桥学院 | Data analysis and early-warning method based on information connotation degree |
CN108150361A (en) * | 2016-12-05 | 2018-06-12 | 北京金风科创风电设备有限公司 | The monitoring method and device of wind generating set engine room vibration |
CN108150361B (en) * | 2016-12-05 | 2019-07-23 | 北京金风科创风电设备有限公司 | The monitoring method and device of wind generating set engine room vibration |
CN110050242A (en) * | 2017-01-26 | 2019-07-23 | 三菱日立电力系统株式会社 | Warning display system and alarm display methods |
CN106960554B (en) * | 2017-03-20 | 2018-06-29 | 山东科技大学 | Alarm system evaluation method and device based on single simulation amount and presignal delay device |
CN106960554A (en) * | 2017-03-20 | 2017-07-18 | 山东科技大学 | Warning system evaluation method and device based on single simulation amount and presignal delay device |
CN108628281A (en) * | 2017-03-23 | 2018-10-09 | 株式会社日立制作所 | Abnormality detection system and method for detecting abnormality |
CN108628281B (en) * | 2017-03-23 | 2021-01-26 | 株式会社日立制作所 | Abnormality detection system and abnormality detection method |
CN109873853A (en) * | 2017-12-05 | 2019-06-11 | 北京绪水互联科技有限公司 | Equipment key parameter early warning system and its implementation, electronic device |
CN109873853B (en) * | 2017-12-05 | 2022-04-15 | 北京绪水互联科技有限公司 | Equipment key parameter early warning system, implementation method thereof and electronic device |
CN108270618A (en) * | 2017-12-30 | 2018-07-10 | 杭州华为数字技术有限公司 | Alert the method, apparatus and warning system of judgement |
CN108270618B (en) * | 2017-12-30 | 2021-07-16 | 华为技术有限公司 | Alarm determination method, device and alarm system |
CN109408303A (en) * | 2018-09-19 | 2019-03-01 | Oppo(重庆)智能科技有限公司 | Test data analysis method and Related product |
CN111626841A (en) * | 2020-04-10 | 2020-09-04 | 中国建设银行股份有限公司 | Method, system and related equipment for monitoring online transaction |
CN111855178A (en) * | 2020-07-23 | 2020-10-30 | 贵州永红航空机械有限责任公司 | Diagnosis method for running state of rotary product |
CN113626906A (en) * | 2021-07-07 | 2021-11-09 | 武汉轻工大学 | Underground supporting structure deformation monitoring method based on data mining and control chart |
CN113626906B (en) * | 2021-07-07 | 2024-03-22 | 武汉轻工大学 | Underground support structure deformation monitoring method based on data mining and control diagram |
CN115913896A (en) * | 2022-11-09 | 2023-04-04 | 中国联合网络通信集团有限公司 | Device detection method, server and medium |
Also Published As
Publication number | Publication date |
---|---|
CN104020724B (en) | 2017-02-08 |
US9798320B2 (en) | 2017-10-24 |
US20140249656A1 (en) | 2014-09-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104020724B (en) | alarm monitoring method and device | |
CN111459778B (en) | Operation and maintenance system abnormal index detection model optimization method, device and storage medium | |
JP5996384B2 (en) | Process monitoring diagnostic device, process monitoring diagnostic program | |
US7194325B2 (en) | System to and method of monitoring condition of process tool | |
JP5091604B2 (en) | Distribution evaluation method, product manufacturing method, distribution evaluation program, and distribution evaluation system | |
DE10394223B4 (en) | Setting a sampling rate based on state estimation results | |
US20050143849A1 (en) | Controlling method for manufacturing process | |
US7050935B1 (en) | Method for assessing the reliability of technical systems | |
US20110161030A1 (en) | Method And Device For Monitoring Measurement Data In Semiconductor Process | |
JP2015011027A (en) | Method for detecting anomalies in time series data | |
CN110874674A (en) | Anomaly detection method, device and equipment | |
CN115293463B (en) | Glass lens processing supervision method and system based on cutting quality prediction | |
US20100060470A1 (en) | Failure cause identifying device and method for identifying failure cause | |
CN104794535B (en) | A kind of method of electric power demand forecasting and early warning based on Dominant Industry | |
CN107077135B (en) | Method and auxiliary system for identifying interference in a device | |
CN111611751B (en) | Chemical process risk dynamic analysis method based on Bayesian and event tree | |
CN105825288B (en) | optimization analysis method for eliminating regression data collinearity problem in complex system | |
CN109100117B (en) | Method for detecting variation value | |
US8340800B2 (en) | Monitoring a process sector in a production facility | |
US10146215B2 (en) | Monitor system and method for semiconductor processes | |
CN105573269B (en) | The parameter monitoring system and method for semiconductor manufacturing board | |
US7855086B2 (en) | Method for monitoring fabrication parameter | |
JP7163117B2 (en) | Process monitoring support device, process monitoring support system, process monitoring support method, and process monitoring support program | |
CN117708615B (en) | Product detection method and processing equipment monitoring method | |
CN116257019B (en) | PLC (programmable logic controller) management method and system based on cloud |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |